This RMarkdown plots the output of the model fits and simulations
## mean se_mean sd 2.5% 25% 50%
## intercept 4.9670415 0.0022611320 0.06857007 4.8408813 4.9165926 4.9698577
## coef 0.6269134 0.0006529733 0.01880611 0.5902723 0.6128387 0.6267878
## sigma_vl 0.8785061 0.0009435690 0.02508941 0.8305398 0.8596759 0.8781823
## t_dof 8.6798023 0.0583875570 1.65201291 6.1009785 7.5227201 8.4578796
## 75% 97.5% n_eff Rhat
## intercept 5.0121482 5.1008207 919.6390 0.9969748
## coef 0.6406164 0.6621930 829.4820 0.9993280
## sigma_vl 0.8959026 0.9276074 707.0231 1.0019773
## t_dof 9.6657497 12.3222168 800.5460 1.0012649
## mean se_mean sd 2.5% 25% 50%
## intercept[1] 1.4210639 0.107346893 0.51792685 0.4731717 1.0568497 1.4170214
## intercept[2] 5.5721391 0.005022655 0.09024346 5.3968523 5.5107993 5.5717195
## coef[1] 0.2050722 0.003779056 0.05182824 0.1085487 0.1670112 0.2029427
## coef[2] 1.0561185 0.015914938 0.09042128 0.9192284 0.9936826 1.0393825
## sigma_vl 0.7202433 0.001014772 0.02719554 0.6657211 0.7027074 0.7209976
## t_dof 5.5235087 0.027741947 0.82251154 4.2157087 4.9470522 5.3926004
## 75% 97.5% n_eff Rhat
## intercept[1] 1.785558 2.3948606 23.27866 1.1024299
## intercept[2] 5.634702 5.7403255 322.82326 1.0113819
## coef[1] 0.238988 0.3099490 188.09023 1.0328515
## coef[2] 1.100133 1.2759119 32.27984 1.1339709
## sigma_vl 0.737933 0.7751848 718.22093 0.9957773
## t_dof 6.028853 7.3935975 879.04301 0.9972616
## There are a total of 280 infection episodes
## [1] 2 18
## [1] 8
There are multiple ways to define time to clearance. We use time to first CT value equal to 40.
mgcv
Sensitivity analysis model 1
Sensitivity analysis model 2
## Warning in predict.gam(mod_gam, data.frame(time = xs, ID = -1), re.effect = NA):
## factor levels -1 not in original fit
Individual fits data 1 - we pick the individuals with the most data
## Doing D max = 5 ....
## Doing D max = 7 ....
## Doing D max = 14 ....
Reduction in AUCs
## Model 1: Wilcoxon test for AUC up to day 5: p= 0.0311334652416703
## Model 2: Wilcoxon test for AUC up to day 5: p= 0.186756104888986
## Model 1: Wilcoxon test for AUC up to day 7: p= 0.0311334652416703
## Model 2: Wilcoxon test for AUC up to day 7: p= 0.163340554379603
## Model 1: Wilcoxon test for AUC up to day 14: p= 0.0351837190994008
## Model 2: Wilcoxon test for AUC up to day 14: p= 0.231141448679665
##
## Welch Two Sample t-test
##
## data: colMeans(thetas_mod1$theta_rand[, not_vacc_ind, 2]) and colMeans(thetas_mod1$theta_rand[, vacc_ind, 2])
## t = -2.6264, df = 33.363, p-value = 0.01293
## alternative hypothesis: true difference in means is not equal to 0
## 95 percent confidence interval:
## -0.3577809 -0.0455068
## sample estimates:
## mean of x mean of y
## -0.10579693 0.09584692
effect of variant
##
## Alpha Delta Epsilon NV
## 14 50 12 27
## Omicron Other OtherVOIVOC Suspected Omicron
## 40 107 5 25
effect of vaccine
##
## 0 1
## 60 17
## effect power
## 1 1.0 0.0210000
## 2 1.3 0.4908750
## 3 1.5 0.7781875
## t_design power
## 1 1 0.676875
## 2 2 0.640750
## 3 3 0.633250
## 4 4 0.587250
## [1] 1000 1000
##
## 5000
## 33